On Thu, Aug 6, 2020 at 12:44 PM Pragnesh Patel
<[email protected]> wrote:
>
> Hi Rick,
>
> >-----Original Message-----
> >From: Rick Chen <[email protected]>
> >Sent: 06 August 2020 08:22
> >To: Pragnesh Patel <[email protected]>
> >Cc: U-Boot Mailing List <[email protected]>; Atish Patra
> ><[email protected]>; Bin Meng <[email protected]>; Anup Patel
> ><[email protected]>; Sagar Kadam <[email protected]>; Paul
> >Walmsley ( Sifive) <[email protected]>; Simon Glass
> ><[email protected]>; [email protected]; [email protected];
> >[email protected]; [email protected]; [email protected];
> >[email protected]; rick <[email protected]>; Alan Kao
> ><[email protected]>
> >Subject: Re: [PATCH] common/board_f: make sure to call fix_fdt() before
> >reserve_fdt()
> >
> >[External Email] Do not click links or attachments unless you recognize the
> >sender and know the content is safe
> >
> >Hi Pragnesh
> >
> >> From: Pragnesh Patel [mailto:[email protected]]
> >> Sent: Wednesday, August 05, 2020 5:01 PM
> >> To: [email protected]; [email protected]; [email protected];
> >> [email protected]; [email protected]; Rick Jian-Zhi Chen(陳建志)
> >> Cc: [email protected]; Pragnesh Patel; Simon Glass; Ovidiu
> >> Panait; Stephen Warren; Patrick Delaunay; Vikas Manocha; Masahiro
> >> Yamada; Ye Li
> >> Subject: [PATCH] common/board_f: make sure to call fix_fdt() before
> >> reserve_fdt()
> >>
> >> There may be a chance that board specific fix_fdt() will change the size
> >> of FDT
> >blob so it's safe to call reserve_fdt() after fix_fdt() otherwise global
> >data (gd) will
> >overwrite with FDT blob values.
> >>
> >> Signed-off-by: Pragnesh Patel <[email protected]>
> >> ---
> >> common/board_f.c | 6 +++---
> >> 1 file changed, 3 insertions(+), 3 deletions(-)
> >>
> >
> >Maybe you can add the fix tag if it is caused by this.
> >Fixes: a8492e25ac71 ("riscv: Expand the DT size before copy reserved memory
> >node")
> >
> >Reviewed-by: Rick Chen <[email protected]>
>
> Good suggestion, will update in v2. Thanks for the review.
I tend to disagree. The ordering issue is there for a long time and
not introduced by a8492e25ac71 so "Fixes" tag is not accurate.
It's just a8492e25ac71 triggered the bug, not introduced the bug.
>
> >
> >> diff --git a/common/board_f.c b/common/board_f.c index
> >> 88ff0424a7..7ae01e9fff 100644
> >> --- a/common/board_f.c
> >> +++ b/common/board_f.c
> >> @@ -956,6 +956,9 @@ static const init_fnc_t init_sequence_f[] = {
> >> * - board info struct
> >> */
> >> setup_dest_addr,
> >> +#ifdef CONFIG_OF_BOARD_FIXUP
> >> + fix_fdt,
> >> +#endif
> >> #ifdef CONFIG_PRAM
> >> reserve_pram,
> >> #endif
> >> @@ -984,9 +987,6 @@ static const init_fnc_t init_sequence_f[] = {
> >> setup_board_part2,
> >> #endif
> >> display_new_sp,
> >> -#ifdef CONFIG_OF_BOARD_FIXUP
> >> - fix_fdt,
> >> -#endif
> >> INIT_FUNC_WATCHDOG_RESET
> >> reloc_fdt,
> >> reloc_bootstage,
Regards,
Bin
